The sooner the better: Use of a real time automated bedside dashboard to improve sepsis care

Authors: Andrew Jung, Michael Goodman, Christopher Droege, Vanessa Nomellini, Jay Johannigman, John B Holcomb, Timothy A Pritts

Presented by University of Cincinnati at the 13th Annual Academic Surgical Congress

Background

Despite advances in modern critical care, sepsis and septic shock remain major contributors to morbidity and mortality. Recent data indicates that decreasing the time interval between the diagnosis and administration of antibiotics is associated with improved patient outcomes. The effect of a visual clinical decision support system on this interval is unknown. We hypothesized that implementation of a commercially available bedside clinical surveillance visualization system would be associated with earlier antibiotic administration and decreased length of stay in surgical intensive care unit (SICU) patients.

Methods

An automated clinical surveillance visualization system was implemented in our SICU beginning in July 2016. This system was integrated with our electronic medical record and continuously displayed vital signs and laboratory data at the bedside on a dedicated clinical dashboard (42” screen). A bedside visual sepsis screen bundle (SSB) was added to the dashboard in June 2017. Among other variables, the clinical dashboard displayed each patient’s calculated sepsis score based on heart rate, body temperature, respiratory rate, and white blood cell count. The SSB indicator turned red if the sepsis score exceeded four points. We retrospectively analyzed prospectively collected data from patients with bedside visualization systems before and after implementation of the SSB. We determined mean sepsis score, maximum sepsis score, time to antibiotic administration, and surgical ICU length of stay.

Results

During the study period, data were collected on 232 patients admitted to the beds with bedside clinical surveillance visualization systems. Of these, 37 patients demonstrated elevated sepsis scores and were given antibiotics for clinical evidence of sepsis or septic shock (26 prior to SSB implementation, 11 after SSB implementation). The mean sepsis score was similar in the pre- and post-SSB groups (1.8 vs 1.6, p=NS) as was the maximum sepsis score (6.0 vs 5.4, p=NS). Time to antibiotic administration was significantly less in the post-SSB patients (pre= 48.9+14 hours vs post= 11.6+6 hours, p<0.05, see figure).  ICU length of stay was also shorter (pre=16.8 days, post=6.2 days, p<0.05) following the introduction of the SSB.

Conclusion

Implementation of a bedside clinical surveillance visualization decision support system was associated with a decreased time interval between the diagnosis of sepsis or septic shock to administration of antibiotics. Integration of decision support systems in the ICU setting may help providers to adhere to Surviving Sepsis guidelines for identification and treatment of surgical patients with infections as well as improve quality of care.

ASC Graph.png